Riven Embed

Embedding sidecar served on Riven's own Blackwell fleet. RAG lane; also reachable via the riven-embed, riven-embed-fast and riven-embed-quality aliases.

qwen3-embedding-0.6b

Input price
$0.14
per 1M tokens
Output price
$0.00
per 1M tokens
Lane
—
Riven lane

Capabilities

embedding

Call Riven Embed in one request

The Riven API is OpenAI-compatible. Point your existing SDK at api.rivenai.io and set the model id — no code changes.

curl https://api.rivenai.io/v1/chat/completions \
  -H "Authorization: Bearer $RIVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "qwen3-embedding-0.6b",
  "messages": [{"role": "user", "content": "Hello"}]
}'

Why teams run Riven Embed on Riven

  • Served from Riven’s own inference fleet — the same lanes that power the Riven apps.
  • One API key and one bill across all 47 lanes in the catalog.
  • Transparent per-token lane pricing, metered from actual fleet cost.
  • OpenAI-compatible endpoint — drop your existing SDK in unchanged.